Assessment and Management of the Flash Floods …3...Saudi Arabia. E-mail: [email protected];...

International Journal of Water Resources and Arid Environments 2(3): 146-157, 2013ISSN 2079-7079© PSIPW, 2013

Corresponding Author: Mohammed El Bastawesy, Department of Geography, Umm Al Qura University, Makkah,Saudi Arabia. E-mail: [email protected]; [email protected].

146

Assessment and Management of the Flash Floods inAl Qaseem Area, Kingdom of Saudi Arabia

Mohammed El Bastawesy and Khalid Al Ghamdi

Department of Geography, Umm Al Qura University, Makkah, Saudi Arabia

Abstract: The available water resources in Saudi Arabia are under increasing stress and the hydrologicalproblem is being complicated by the adoption of inadequate management strategies. The hydrologicalparameters (drainage networks, flow direction and flow length) have been extracted from the DEM of the SRTMusing the automatic procedures embedded into the ArcGIS algorithms. The multi-temporal satellite images ofthe TM and ETM+ were obtained and processed to delineate the landcover and also to detect the extent of wadialluvium, active channels and the accumulated water ponds. Cross sectional areas of the active channels andthe extracted hydrological parameters of the DEM were integrated into the open channel flow equation ofManning’s to determine the spatially distributed time area zones of Wadi Al Rimah. The measuring of changesin cross-sectional areas of an active channel reach free of any tributary inflow and the estimated flow travel timegave a transmission loss rate of 12 mm per hour. The TRMM data of the 7th of January 1999, the time area zonesand the estimated transmission loss were used to compute the runoff hydrograph. GIS analysis of the extentof resulting water ponds and bottom topography of the playas dotting the Mesozoic carbonate rocks at outlet,showed that the excess runoff is approximately 19% and the flash flood event of January 1998 has deliveredabout 252 million cubic meters to the main playas. Outcropping of the Paleozoic sandstone (i.e. Al Saq aquifer)at the lower part of the catchment provides a good opportunity to harvest these occasional flows otherwiserepresenting a hazard and being wasted. The construction of a dam downstream is quite feasible, but a detailedgeotechnical and environmental impact assessments are required.

Key words: Remote sensing GIS Flash floods Management Transmission loss Wadi Groundwater

INTRODUCTION prone to flash floods [2]. Flash flooding and associated

The drylands are suffering from scarcity of water environmental problem, due to the relative infrequentresources, with the exception of small particular areas occurrence of runoff events and lack of observationsendowed by fossil groundwater or the collection of runoff [3, 4]. The absence of good quality data of rainfall,discharge. Currently, much of the non-renewable deep discharge and sediment erosion may present a particularaquifers are being depleted by the aggressive pumping of problem for hydrological modeling in the drylandfossil groundwater for irrigation as well as other different catchments [5].uses. Therefore, the fragile resources of the inhabited This poor understanding of the hydrologicaldryland areas are under increasing pressure of population processes has further been complicated by the interactiongrowth and urban expansion, which in turn resulted in of urban development and resulting runoff. Urbanizationsevere environmental degradation [1]. Consequently, the process itself through the construction of imperviousrapid deterioration and depletions of water resources of surfaces; building, roads, storm sewers and paving hasthe dryland can threaten the local environment and decreased the infiltration capacities of the underlyinginadequate water supplies which may create crises in the soils and it significantly increase runoff-dischargenear future. On the other hand, urban areas and of course downstream [6, 7]. Therefore, the runoff could bethe agricultural areas are favorably developed on the contaminated by different types of pollutants dependingexpense of alluvial plains and dry channel beds, which are on activities and waste products from urban areas.

sediment transport in drylands are often overlooked as an

Intl. J. Water Resources & Arid Environ., 2(3): 146-157, 2013

147

Moreover, site selection of the adopted management to identify and map multi-temporal land use-land coverstrategies for flash floods (e.g. dams) could only fulfill the features [12], which in turn can be linked with dominantrequired protection, but the collected runoff itself is processes, activities and microscopic observations towasted. Therefore, it is so necessary to understand the identify the hydrological problems. The analysis ofinterplay of geo-environmental setting and development multi-temporal satellite images particularly those acquiredvariables to determine the required measures, which have shortly before and after flash flood events are useful toto be adopted to achieve efficient management of fragile identify the active channels and to infer important runoffwater resources. parameters such as maximum channels -wetted perimeters

The management strategies of flash floods are (i.e. peak flows), water ponds and estimates ofcontrolled by the hydrogeological and climatological transmission losses [5]. On the other hand, the mappingcharacteristics of the catchments, land use pattern and of drainage channels and catchment divides can bedistribution and the local water use and need. Indeed, the derived with reasonable accuracy using the availablerainfall pattern in drylands presents the main difficulty for DEM for most of the globe such as, the Shuttle Radara useful runoff harvesting. In total, the precipitation is Topography Mission (SRTM) DEM of 90 m resolutionerratic and rainfall events can be spaced over years; thus [5] and also from the Advanced Spaceborn Thermalthe recharge to groundwater usually is very low when Emission and Reflection Radiometer (ASTER) DEM ofcompared with the mean annual precipitation [8]. 30 m resolution [13]. Whatever the source of drainageHowever, the exceptional great storms occasionally networks and utilized processing algorithms, theoverpass the total annual precipitation; can thus provide comparison of these derivatives against their counterpartsa good opportunity for considerable groundwater on satellite images or detailed field surveyed topographicrecharge as the accumulated runoff is abundant [9]. maps remains a necessary measure to assess the quality

The impoundments and runoff harvesting are key of drainage portraying and the accuracy of consequentelements in the management of the dryland drainage hydrological analysis [14].basins ranging from micro catchments covering small The aim of this paper is to investigate the hydrologyareas (i.e. few hectares) to mega catchments draining of Wadi Al Rimah draining Al Qaseem area in Saudiseveral hundred square kilometers [10]. Several Arabia. The flash floods will be quantitively assessed intechniques and structures are being implemented in order to determine the current negative impacts and alsodifferent Saharan areas including, but not limited to; to determine the potential to the development of limitedexcavated and rock-cut cisterns, contour trenching and water resources. The available multi-temporal satelliteterracing, retention dykes for water ponding and farming, images and DEM and geological data will be analyzed toearth-core structures and dams constructions. Despite the select the optimum locations suitable for harvestingfluctuations in runoff; the sedimentation problem runoff and replenishing the main aquifer of Al Saq(particularly for reservoirs of dams) is not only reducing supporting the agricultural activates in the area.the storage capacities, but also affects the conditions ofgroundwater recharge and the stability of these Study Area: The agricultural areas are being distributedstructures. The management of flash floods and runoff throughout the Kingdom of Saudi Arabia, particularly inharvesting can only be improved if the hydrological data the desert belts underlain by groundwater aquifers. It isare sufficient and the processes of the catchments are estimated that 1.5 million hectares of land are beingwell-understood and fully considered [11]. This is cultivated, mainly on the low laying soils of playa andbecause the imperfect knowledge of the characteristics alluvial plains in different drainage basins within theof such flows, compounded by the scarcity of relevant Kingdom [15]. Although concerns are growing for thedata and observations, makes it extremely difficult to aggressive pumping of the groundwater aquifers, whichevaluate the potential magnitude and frequencies of are believed to depleted, additional groundwater-fedflows. cultivations are being developed. Al Qaseem area, which

The lack of data problem has considerably been is located approximately in the centre of Saudi Arabia,overcome as megascopic observations and various gains its importance as it contains more than 50% of thequantitative data can be obtained and extracted from the total cultivated areas in the Kingdom and drained by largewidely available satellite images and Digital Elevation drainage basins, which have their headwaters carved inModels (DEM) of different sources and resolutions [5]. the Asir mountain series to the west (Fig. 1). Generally,Straightforward, remote sensing images are widely used rainfall in the Kingdom takes place primarily during winter


148

Fig. 1: Location map of the study area, the solid rose areas refer to locations of Figure 3

and spring, which are caused by combination of great curved belt along the eastern edge of the olderdisturbances from the Mediterranean and the Sudan crystalline rocks. The basal sequence of thesetrough [16]. Annual rainfall of 100 to 200 mm occurs in the sedimentary rocks (i.e. Al Saq Formation) are mainlycentral area extending from north of Riyadh and Al composed of sandstone strata and represent the principalQaseem to the vicinity of Hail [17]. The distribution of aquifer in the central part of the Kingdom. The age ofrainfall in the area is extremely variable; rainfall is sporadic Al Saq sandstone and its equivalent units (i.e. Wajidand erratic. The prevailing arid conditions are sandstone in southwestern Saudi Arabia) ranges fromoccasionally interrupted by some heavy storms that Cambrian to Ordovician, based on fossil contents as wellgenerate channel-runoff, which could confluence and as elemental chemistry [19]. Al Saq sandstone followsattenuate downstream into destructive flash floods. closely the regional eastward dip of the Pre-CambrianThe city of Buraidah (i.e. the main city in Al Qaseem basement; the depth to the sandstone below the groundregion) and its neighboring rural and cultivated areas are surface could reach 2000 m at a distance of 130 km east ofbeing affected by these severe flash floods mainly the outcrop. Boreholes show that the regional dip isproduced by Wadi Al Rimah. The catchment of Wadi exceptionally steep in Tabuk area in the northwest ofAl Rimah equates an area of 107,000 sq km and it is Saudi Arabia. The thickness of Al Saq sandstone alsoconsidered one of the biggest wadis that are draining the varies from 400 in the southern part of Al Qaseem toAsir-Najd plateau eastward. 800 m in the northern part of the study area [20].

The upper reaches of the wadi system are carved in The uncontrolled withdrawal from the main aquifersthe resistant Precambrian igneous and metamorphic rocks with little attention paid to the volume of water actuallybelonging to the Arabian Shield and they are mainly required for irrigation, considerable water have beenoutcropping in the western part of the catchment (Fig. 2). wasted from the agricultural drainage system. By 1984,The eastern edge of these basement rocks are flanked some artesian wells had been in use for nearly 20 yearsby some lava fields (i.e. Harrat) developed during the have ceased and the water table in certain areas isMid-Tertiary and Quaternary periods [18]. Generally, considerably being reduced [21]. The rapid decline ofthe basaltic flows of Harrat are being dotted by salt water table indicates that the recharge is almostflats occupying the lava cones and the abandoned negligible and is occurring only during heavy flash floodpaleochannels. The Paleozoic sedimentary rocks form a events affecting these outcrops. The annual recharge is


149

Fig. 2: The landcover map of Wadi Al Rimah, based on the geological maps, DEM and Landsat TM image of 1999

estimated in Al Qaseem area to 80 million cubic meters of availability of remotely sensed data including orbitalwater, while in Hail to the north the annual recharge is measurements of rainfall such as the Tropical Rainfallestimated to 20 million cubic meters of water [21]. The Monitoring Mission (TRMM) and the monitoring ofsamples collected from wells drilled in Al Saq aquifer landuse and landcover changes through differentshowed a radiocarbon age of 28,000 years, which imply operating satellites such as Landsat Thematic Mapperthe significance of the Quaternary wet climatic periods on (TM) and Enhanced Thematic Mapper (ETM+). Thethe hydrology of Saharan deserts in Saudi Arabia [22]. quantitative assessment of a flash flood (e.g. duration,

Therefore, it is necessary to investigate distribution peak discharge rate and total flow volumes) resulting fromof the aquifer-outcrops in the catchment and the given storm required the integration of several datasetshydrology of flash floods to determine the potential including TRMM data, SRTM DEM and multiple Landsatharvesting of hazardous runoff otherwise being wasted. images particularly those acquired shortly before and afterSeveral severe flash floods were recorded during the past the flash floods. The open channel flow equation offew decades and were capable to reach Buraidah at the Manning’s was used to estimate the runoff velocities atcatchment; and the negative impact of these flows was certain cross sectional areas of the active channels inconsiderable. This clearly indicates that these flows can order to determine the time-consumed by flows to reachbe controlled particularly at the outcrop area of Al Saq the outlet. Therefore, several processing procedures wereand thus increasing recharge to the declining executed to firstly determine the flash flood hazard on thegroundwater. developed areas of Wadi Al Rimah and secondly to

MATERIALS AND METHODS areas of the catchment to increase the recharge of

The management of flash floods in the dryland The Landsat satellite images of March 1987, Februaryenvironment is confronted by the availability of limited 1998, April 1998, July, 1998 and January 1999 weredatasets on rainfall, transmission losses, runoff, land processed and interpreted to delineate the active channelscover and hydrological structures in the area. This and water ponds accumulated in the playas dotting thelimitation is partially being overcome by the increasing lower part of the catchment (Fig. 3). The TRMM image for

estimate the potential harvesting of flash floods at certain

groundwater.


150

Fig. 3: Distribution and extent of active channels and flash floods based on the Landsat images, March 1987 (top),February 1998 (middle) and January 1999 (bottom)

0.67 0.5R SVn

=


151

Fig. 4: DEM of Wadi Al Rimah catchment

the flash flood events of January 1998 and January 1999 Once the channel networks were extracted, thewere obtained and processed into GIS to determine the spatially distributed time area zones were calculated indistribution of rainfall depths over the entire catchment. order to calculate the hydrograph at certain outlets and toThe DEM (Fig. 4) has been used to extract the drainage determine the different flow parameters for the crossnetworks and to calculate the different time-area zones sectional areas indicated by recent active channels on theapplying the open channel flow equation of Manning’s. satellite images. However the resolution of SRTM is notTheses analyses are useful to investigate the spatial high enough to represent an accurate cross section (i.e. itrelationship of agricultural fields and the land cover units does not capture the small scale terraces or leaves withinwith the drainage networks and also to estimate the runoff the channels). But most open channel methods requirehydrographs using the input rainfall data, time area zones that channel cross sections are generalised prior toand estimates of transmission losses. estimating flow characteristics [25], so SRTM data should

The hydrological analyses of DEM were carried out provide a reasonable approximation for the relatively largeusing the widely used D-8 algorithm embedded in ArcInfo channels considered here.software with some modifications in the filling of DEM The overland and channel flow velocity wasstep [23]. This method requires, first, that all the sinks (i.e. estimated empirically using the Manning Equation:local depressions) of the DEM to be filled and raised inelevation to their neighbouring cells in order to ensure the (1)flow continuity within the catchment to an outlet [24].Then the flow direction of each cell into the lowest where:elevation cell of the surrounding eight cells was V : Is the cross-sectional average velocity (m/s)determined for the entire DEM. Once the route of flow is n : Is the Manning coeffiecent of roughness determined for each cell, it is possible to accumulate the R : Is the hydraulic radius (m)number of upslope flow contributing cells (i.e. areas) S : Is the slope of the water surface which is assumedand the flow pathways. Finally a threshold of the minimum to be parallel to the slope of the channel bed.flow accumulation number was set to extract the fingertipsof delineated channels and different catchments within The widths of the channels at selected cross sectionsthe area. were measured from the active channels mapped from the


152

satellite images. These selected cross sections were channels and the other events were implied by markedoverlaid on the DEM for the estimation of mean depth. increase in brightness of channel bed materials within theSlope was estimated from the DEM for reaches centred on active channels and the accumulation of water ponds inthe selected cross sections. Hydraulic Radius (which is low lying playas. The different parts of the catchment didnearly equal to the depth of flow in shallow braided not equally contribute runoff to outlet as indicated by thechannel systems in dryland areas [3] was calculated from reflectance of channels beds. This could be related to thethe measured cross section areas and perimeter. spatial pattern of rainfall events over the area. ForManning’s n was set to 0.02 for channels and 0.06 for example, GIS analysis of TRMM data of the 5 of Januaryhillslopes; these values are typical of reported values from 1998 showed that the storm has covered the lower half ofsimilar catchments [26]. Once the flow velocity and flow the catchment having its centre over the outlet, whereaslength within each cell are calculated, it is possible to the storm of 7 of January 1999 has nearly affected theestimate the travel time for every cell in the catchment. entire catchment and the highest intensities were recordedTherefore, the catchment can be subdivided into different over the western and southern parts of the catchmenttime-area zones (in hours) separated by isochrones. (Fig. 5).

This time-area diagram for the catchment can However, the local geology and land cover may be ofrepresent a spatially distributed unit hydrograph without importance here; the low undulating topography ofthe need for empirical functions for the time of eastern tributaries are covered by Quaternary sand sheetsconcentration. The resulting rainfall/runoff of each time and alluvium. The runoff initiation threshold for thesezone can be routed downstream in a cascading manner areas is likely to be high, as infiltration rates are muchthat considers abstractions due to storage or loss over higher due to the porous nature of the substrate.the segments of flow pathway. Therefore, the runoff contributing areas, even under the

RESULTS Wadi Al Rimah is ideal to examine the capabilities of

Significant runoff events in the past two decades in quantitatively estimate the accumulating flow volumes; asthe catchment of Wadi Al Rimah were monitored using the catchment is closed and the playas are occupying theseveral sets of available satellite data. Some of these outlet. There was also a good correlation between theevents were coincidently captured while flows were in the TRMM rainfall data and the extent of active channels as

th

th

same rainfall amounts, will be different. The catchment of

remotely sensed data, GIS and DEM derivatives to

Fig. 5 A: TRMM estimated total rainfall (mm) and distribution over Wadi Al Rimah catchment on the 7 of January 1999th

Note: the correspondence between the rainfall pattern and the resulting active channels (red lines), which are digitizedfrom the Landsat image acquired shortly after this rainfall event (Fig. 3)


153

Fig. 5 B: TRMM data of the 5 of January 1998 rainfall eventth

Note the considerable differences in rainfall patterns each event

Fig. 6: The resultant water ponds of the flash flood of January 1998Note: the changes of surface extent between February and April 1998. The stored water volumes in ponds 1, 2 and 3 werecomputed using these delineated surface areas and the dry-bottom topography derived from the SRTM

well as the resulting water ponds on the satellite images. were digitized from the satellite images of these eventsTherefore, the significant rainfall and runoff accumulation and the bottom topography of the inundated playascan be detected by different remote sensing platforms and (i.e. depth) was extracted from the SRTM to compute thecomputed in GIS environment. stored water volumes. The estimated total volume of the

The spatial distribution and rainfall depths were used rainfall of 5 of January 1998 is approximately 1274 millionto estimate the total rainfall volumes over the entire cubic meters of water and the resulting runoff watercatchment. Thereafter, the aerial extent of water ponds delivered to three main playas; namely 1, 2 and 3 ( Fig. 6)

th


154

Fig. 7: The simulated runoff hydrograph for Wadi Al Rimah using the entire flow contributing areas (i.e. time zones) andthe distributed rainfall of Fig. 5 A

Note: the impact of transmission loss on the hydrograph peaks and attenuations

has covered surface areas of 26, 19 and 15 km and dam quite feasible, as the main channel reaches 1.6 km in2

measured 93, 57 and 92 million cubic meters of water average and the surrounding rock shoulders are 19 mrespectively. On average, the resulting runoff that could higher than the channel bed.be harvested may reach 19% of the significant rainfallevents, which affects large portion of the catchment and DISCUSSIONattains a considerable depth and intensity.

The spatially distributed time-area data suggest that Magnitudes of flash floods are controlled by athe flow generated at the most western upstream combination of rainfall characteristics (amount, intensity,hillslopes requires a maximum of 71 hours to reach the duration and distribution over the catchment).outlet (Fig. 7). The most significant finding is that most of Additionally, local lithological and geomorphologicalthe runoff infiltrates into the alluvium beneath the channel conditions also influence the runoff production within thebed and bank during its transmission through the catchment, even under the same meteorologicalchannel. During the January 1999 event, the estimated conditions. In general, the whole catchment is unlikely toflow travel time from a certain channel reach, free from any be uniformly affected by an individual storm. Therefore,tributary flow until it get totally infiltrated was the flow-producting hillslope areas will be different fromapproximately 6 hours. Most of runoff within this tributary one storm to another. This temporal and spatial variationchannel has completely been lost before reaching the in runoff processes makes it extremely difficult tooutlet of the catchment. The peak discharges were implied generalise a standard conceptual model for drylandby the estimating wetted perimeter of the active channel catchments. This has given rise to a plethora ofat spaced apart- points. This estimation gave a hydrological models which incorporate different elements.transmission loss rate of 12 mm\hour for the flow to get It is entirely possible that more than one rainfall event hastotally infiltrated by reaching the most distant location occurred between acquisition of the successive imagesdownstream. This obtained estimate of transmission loss used for this research. This could result in errors inrate was considered to calculate the hydrograph of the discriminating the contributing hillslopes and estimatingflash flood event of 7 of January 1999, which could not the potential runoff to be harvested and recharged intoth

be calibrated simply because the gauging of runoff is not Al Saq aquifer. The location of outcrops of Al Saq aquiferavailable. The outcropping of Al Saq aquifer in the at the lower part of the catchment makes the potentialeastern part of the catchment provide a good opportunity harvesting of runoff only significant during major flashto control the flash floods upstream of Buraidah and to floods of long return periods. This is becausecollected these flows to be recharged into the main transmission loss significantly reduces the volumes ofaquifer, which is suffering from aggressive pumping. runoff. The majority of runoff events produced upstream,The topography of Al Saq peniplain in the downstream particularly in dryland catchments of thousands kmpart of the catchment makes the construction of proposed surface area, will not reach the outlet. Many flash flood

2


155

models ignore this critical process and overestimate had constructed upto 100 dams of various types and sizesrunoff because they model the hydrograph as simply the in different drainage basins. Two of the largest structuresgross runoff production from the different land cover constructed upto that time were the Wadi Jizan Dam andunits within the catchments [2, 4, 27]. The overland flow the Mudhiq Dam in Najran, which have storage capacitiesproduction is lumped from the whole catchment and the of 71 and 86 million cubic meters, respectively [21]. Mostrole of channel networks, underlying rock units, geometry of these modern storage dams are located in the upstreamand pattern is too often overlooked. Realistically, dryland of mountainous area of southwestern Saudi Arabia,hydrological modelling approach should be semi- whereas most of the small recharge dams are build in thedistributed, not just for overland flow production, but also Najd region. Additionally, the operation of these damsfor the subsequent channel routing and transmission depends on releasing the floodwater thorough gatedloss. Such models can be implemented using existing GIS outlets, which are provided at different levels to regulatetechniques, as outlined in this paper. Although, the the rate of outflow for irrigation and domestic use andsimulated hydrographs whether produced for the whole also to recharge the alluvial aquifer downstream.or part of the catchment (Fig. 7) can not be tested, However, the groundwater resources have been depletedbecause there are no flow observations. This is why the from some alluvial aquifers, where dams were constructedmethod has been developed and this research has the upstream on the fingertips channels incised in ruggedpotential of investigating the multi-temporal satellite mountainous area and the annual recharge rate is very lowimages to define the occurrence of floods in arid and and varies between 3 and 21 % of the total annualhyper-arid areas where installing and maintaining participation [8]. Consequently, the collection ofinstrumentation is difficult. floodwater into reservoirs upstream has deprived the thin

The active channel cross sectional areas were alluvial aquifers in most downstream areas fromestimated from these images and used as input to a simple replenishment by the occasional flows. Furthermore, theopen-channel flow equation (Manning) which has been non-regulated pumping of groundwater upstream is alsowidely used for both overland and channel flow affecting the groundwater flow into downstream area ofestimation for both dryland and humid catchments the aquifer used to sustain the rural areas. The[28-30]. The isochron zones of the catchment [31] degradation of groundwater in Wadi Fatmah in Makkahwere delineated and used to model flow propagation area is a clear example; the agricultural production fromthrough the catchment. The accuracy and resolution of the fields in Wadi Fatmah has mostly sustainedthe DEM [32, 33] and other hydrological parameters Makkah since early historical periods until few decades(e.g. Manning’s n) have a significant effect on the ago. The groundwater-fed cultivations of Wadi Fatmahcatchment hydrograph, but the sensitivity of model have almost disappeared following the aggressiveoutput to these parameters is addressed elsewhere [26]. pumping of its fragile groundwater resources, whichAs part of the next phase of research, the method started in 1947 to support the municipal water supplies todeveloped in this study to estimate transmission losses Makkah as well as Geddah (80 km west of Makkah on thecan be applied to instrumented dryland-catchment for Red Sea coast) [35]. Therefore, the aquifer is almostvalidation. Where the satellite images acquired shortly depleted and most of the fields are abandoned, as thebefore and after the occurrence of key flash flood events sporadic and erratic rainfalls are barely sufficient toalong with the DEM can be used to simulate the runoff replenish the underlying aquifer systems of rural areashydrographs. upstream.

It is of utmost importance to understand the responseof dryland catchments to rainfall events is essential in CONCLUSIONorder to assess the impact of flash floods on theincreasing anthropogenic activities within these The lack of hydrological data for the drylandcatchments. Furthermore, the groundwater resources are catchments affects the management strategies beingbeing withdrawn at rates much larger than the recharge implemented for controlling flash floods and harvestingoccurring during these flash floods. Therefore, the runoff. This data limitation can be significantly overcomemanagement of flash floods should consider both by the analysis of active channels affected by recent flashprotection and maximizing the benefits of produced floods. These channels are marked by distinctiverunoff. Dams have been constructed throughout Saudi reflectance on the satellite imageries such as Landsat TMArabia from antiquity for water storage, flood control and and ETM+. These data are used, along with DEMirrigation [34]. Until 1983 the Kingdom of Saudi Arabia derivatives, in a GIS for the determination of synthetic


156

spatially distributed time area–zones of Wadi Al Rimah 9. Qadir, M., B.R. Sharma, A. Bruggeman, R. Choukr-catchment; thus estimating the flash flood pathways and Allah and F. Karajeh, 2007. Non-conventional waterdischarge and also to estimate potential runoff to be resources and opportunities for water augmentationharvested. It is highly recommended to construct a large to achieve food security in water scarce countries.storage dam at the lower part of Wadi Al Rimah, where Agricultural Water Management, 87: 2-22.Al Saq sandstone is outcropping by the main channel. It 10. Lange, J., C.H. Leibundgut, N. Greenbaum andis important to recognise that produced runoff within A.P. Schick, 1999. A noncalibrated rainfall-runofflarger catchments can be completely lost into the model for large, arid catchments. Water Resourcesunderlying channel alluvium before reaching the outlet. Research, 35: 2161-2172.Therefore, this proposed dam will significantly contribute 11. Simmers, I., 2003. Hydrological processes and waterto recharging Al Saq aquifer from the severe flash flood of resources management. In: I. Simmers, ed.long return periods otherwise capable to reach Buraidah. Understanding Water in a Dry Environment:Finally it is recommend that dryland hydrological Hydrological Processes in Arid and Semi-arid Zones,modelling should take advantage of the increasing Balkema Publishers, Chapter, 1: 1-14.availability of high spatial and temporal resolution satellite 12. Schultz, G.A., 1987. Parameter determination andimages, so that flash flood protection measures can be input estimation in rainfall-runoff modeling basedproperly managed and allocated. This will also maximise on remote sensing techniques. Water for thethe benefits of the limited water resources available in the Future: Hydrology in Perspective (Proceedings ofdryland. the Rome Symposium, April 1987). IAHS Publication,

REFERENCES 13. Yang, L., X. Meng and X. Zhang, 2011. SRTM DEM

1. Sorman, A.U. and M.J. Abdulrazzak, 1993. Infiltration- of Remote Sensing, 32: 3875-3896.recharge through wadi beds in arid regions. 14. Wise, S.M., 2000. Assessing the quality forHydrological Science Journal, 38: 173-186. hydrological applications of digital elevation

2. Foody, G.M., E.M. Ghoneim and N.W. Arnell, 2004. models derived from contours. HydrologicalPredicting locations sensitive to flash flooding in an Processes, 14: 1909-1929.arid environment. Journal of Hydrology, 292: 48-58. 15. Al Harbi, K.M., 2010. Monitoring of agricultural area

3. White, K., 1995. Field techniques for estimating trend in Tabuk region-Saudi Arabia using Landsatdownstream changes in discharge of gravel-bedded TM and SPOT data. The Egyptian Journal of Remoteephemeral streams: a case study in southern Tunisia. Sensing and Space Sciences, 13: 37-42.Journal of Arid Environments, 30: 283-294. 16. Almazroui, M., 2011. Calibration of TRMM rainfall

4. Gheith, H. and M. Sultan, 2002. Construction of a climatology over Saudi Arabia during 1998-2009.hydrologic model for the estimating wadi runoff and Atmospheric Research, 99: 400-414.groundwater recharge in the Eastern Desert, Egypt. 17. Ministry of Agriculture and Water, 1988. ClimateJournal of Hydrology, 263: 36-55. Atlas of Saudi Arabia, pp: 118.

5. El Bastawesy, M., K. White and A. Nasr, 2009. 18. Al Ahmadi, M.E., 2009. Hydrogeology of the SaqIntegration of remote sensing and GIS for modeling aquifer, northwest of Tabuk, northern Saudi Arabia.flash floods in Wadi Hudain catchment, Egypt. Journal of King AbdulAziz University: EarthHydrological Processes, 23: 1359-1368. Sciences, 20: 51-66.

6. Leveson, D., 1980. Geology and the urban 19. Hussain, M., 2007. Elemental chemistry as a tool ofenvironment. Oxford University Press, New York, stratigraphic correlation: A case study involvingpp: 386. lower Paleozoic Wajid, Saq and Qasim formations

7. Shang, J. and J.P. Wilson, 2009. Watershed in Saudi Arabia. Marine and Petroleum Geology,urbanization and changing flood behavior across the 24: 91-108.Los Angeles metropolitan region. Natural Hazards, 20. Lloyd, J.W. and R.H. Pim, 1990. The hydrogeology48: 41-57. and groundwater resources development of the

8. Bazuhair, A.S. and W.W. Wood, 1996. Chloride mass- Cambro-Ordovician sandstone aquifer in Saudibalance method for estimating ground water recharge Arabia and Jordan. Journal of Hydrology, 121: 1-20.in arid areas: example from western Saudi Arabia. 21. Ministry of Agriculture and Water, 1984. Water AtlasJournal of Hydrology, 186: 153-159. of Saudi Arabia, pp: 112.

pp: 164.

and its application advances. International Journal


157

22. Sowayan, A.M. and R. Allayla, 1989. Origin of the 29 Rico, M., G. Benito and A. Barnolas, 2001. Combinedsaline groundwater in the Wadi Ar-Rumah, Saudi palaeoflood and rainfall-runoff assessment ofArabia. Ground Water, 27: 481-490. mountain floods (Spanish Pyrenees). International

23. El Bastawesy, M., R.R. Ali and A.H. Nasr, 2008. The Journal of Hydrology, 245: 59-72.Use of remote sensing and GIS for catchments 30. Liu, Y.B., S. Gebremeskel, F. De Smedt, L. Hoffmann,delineation in Northwestern Coast of Egypt: An and L. Pfister, 2003. A diffusive transport approachassessment of water resources and soil potential. for flow routing in GIS-based flood modelling.The Egyptian Journal of Remote Sensing and Space International Journal of Hydrology, 283: 91-106.Sciences, 10: 1-13. 31. Maidment, D.R., 1993. HydroGIS 93: application of

24. Jenson, S.K. and J.O. Domingue, 1988. Extracting geographical information system in hydrology andtopographic structure from digital elevation data for water resources. Proceeding of the Viennageographical information system analysis. Conference, IAHS Pub. No. 211.Photogrammetric Engineering and Remote Sensing, 32. Wolock, D.M. and C.V. Price, 1994. Effects of digital54: 1593-1600. elevation map scale and data resolution on a

25. Hey, R.D., 1979. Flow resistance in gravel-bed topographically based watershed model. Waterrivers. Journal of the Hydraulics Division, HY4, Resources Research, 30: 3041-3052.105: 365-379. 33. Zhang, W. and D.R. Montgomery, 1994. Digital

26. El Bastawesy, M., 2006. Influence of DEM source and elevation model grid size, landscape representationresolution on the hydrographical simulation of and hydrological simulations. Water ResourcesWadi Keed catchment, Southern Sinia, Egypt. The Research, 30: 1019-1028.Egyptian Journal of Remote Sensing and Space 34. Makki, M.S., 1979. Dams around the Holy City ofScience, 9: 68-79. Madina and the pressure on its water supply. Journal

27. Costa, J.E., 1987. Hydraulics and basin morphometry of Arid Environments, 2: 363-367.of the largest flash floods in the conterminous United 35. El Baroudi, M.S., 1986. The water balance model forStates. Journal of Hydrology, 93: 313-338. Wadi Fatima. The Journal of Geographical Society of

28. O'Connor, J.E. and R.H. Webb, 1988. `Hydraulic Kuwait, 88: 1-63. ( in Arabic).modelling for palaeofood analysis', In: V.R. Baker,R.C. Kochel and P.C. Patton, (Eds), FloodGeomorphology. Wiley Interscience, New York,pp: 383-402.

Assessment and Management of the Flash Floods …3...Saudi Arabia. E-mail: [email protected];...

Documents

Transcript of Assessment and Management of the Flash Floods …3...Saudi Arabia. E-mail: [email protected];...